Development of eptifibatide

Interventional Cardiology

Development of eptifibatide Robert M. Scarborough, PhD South San Francisco, Calif

Background The primary cause of acute coronary syndromes is the development of a thrombus, a pathologic manifestation of platelet aggregation that occurs as part of the normal process of hemostasis. The discovery that the final common step in platelet aggregation, through the binding of fibrinogen to the activated platelet integrin glycoprotein (GP) IIb/IIIa, has opened the door to the development of novel and potentially more effective antithrombotic therapies. Abciximab, a human-murine chimeric Fab fragment of a monoclonal antibody against the GP IIb/IIIa receptor, was the first agent of this class to demonstrate clinical effectiveness. Several of the specific properties of abciximab, such as its long half-life, lack of receptor-blocking specificity, and some tendency for antigenicity, have prompted the development of alternative GP IIb/IIIa inhibitors with distinct pharmacologic profiles.

Methods and Results One of these newer agents is eptifibatide, which was developed by mimicking the GP IIb/IIIa blocker barbourin, found in the venom of the southeastern pigmy rattlesnake. Eptifibatide is a small, cyclic heptapeptide that has shown high specificity and high affinity for GP IIb-IIIa, a short plasma half-life, and rapid onset of antiplatelet action accompanied by a rapid reversibility of platelet inhibition once treatment is stopped. Conclusions In clinical trials, culminating in the phase III IMPACT II (Integrilin to Minimize Platelet Aggregation and Coronary Thrombosis) and PURSUIT (Platelet GP IIb-IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy) trials, eptifibatide was found to reduce coronary events significantly in a broad range of low-, medium-, and high-risk patients with acute coronary syndromes without significantly increasing the risk of bleeding or other complications. These results suggest that eptifibatide may prove to be an effective addition to currently available antithrombotic therapies. (Am Heart J 1999;138:1093-104.)

When a normal blood vessel is damaged, disruption of the endothelial lining of the vessel wall triggers a cascade of responses that minimize blood loss if the injury is severe and initiates mechanisms of vessel repair. Hemostasis, the spontaneous arrest of bleeding, is the crucial event in this process and occurs within seconds of injury. It is characterized by the adhesion of circulating platelets to subendothelial collagen or other adhesive proteins that have been exposed by the injury followed by platelet activation and aggregation. Aggregated platelets create a hemostatic “plug” that provides the actual physical barrier to blood loss at the site of injury. Completing the process of hemostasis, in part as a direct consequence of platelet activation, the coagulation cascade is stimulated, resulting in production of thrombin and the local deposition of fibrin.1 Thrombus formation (thrombosis) on a fissured or disrupted atherosclerotic plaque can be understood as a pathologic manifestation of the normal process of From COR Therapeutics, Inc. Guest Editor for this manuscript was David P. Faxon, MD, University of Southern California School of Medicine, Los Angeles. Submitted August 6, 1998; accepted December 10, 1998. Reprint requests: Robert M. Scarborough, PhD, COR Therapeutics, Inc, 256 E Grand Ave, South San Francisco, CA 94080. Copyright © 1999 by Mosby, Inc. 0002-8703/99/$8.00 + 0 4/1/96757

hemostasis. The thrombus, a large platelet- or fibrin-rich plug that may impede or completely block blood flow through the affected vessel, is triggered by plaque rupture, resulting in exposure of thrombogenic stimuli within the plaque and in the subendothelium. As in hemostasis, these stimuli lead to platelet adhesion, activation, and aggregation and to activation of the coagulation cascade with fibrin deposition. The resulting plug, or thrombus, slows or prevents the flow of blood through the coronary vessel, causing myocardial ischemia.1,2 Therefore coronary blood vessel thrombosis is the primary pathologic mechanism associated with the clinical presentation of acute coronary syndromes (ACS), which include unstable angina, myocardial infarction (MI), and sudden ischemic death. The symptoms caused by formation of intracoronary thrombus will depend on the extent and duration of the thrombotic occlusion. Mural, nonocclusive, white thrombi, consisting predominantly of aggregated platelets, are the cause of cardiac ischemia in most patients (>90%) with unstable angina.2,3 The pathophysiologic characteristics of nonQ-wave MI are similar to those of unstable angina, but this syndrome is accompanied by more severe ischemia and evidence of myocardial necrosis.1,4 White thrombi also represent an initial stage in the formation of arter-

1094 Scarborough

American Heart Journal December 1999

Figure 1

Role of GP IIb/IIIa in platelet aggregation. On unstimulated platelets, GP IIb/IIIa is in a conformation that has low affinity for soluble fibrinogen. When platelets are activated, they undergo morphologic and physiologic changes, and GP IIb/IIIa molecule alters its conformation, becoming a high-affinity receptor for fibrinogen. Each fibrinogen molecule can bind to 2 GP IIb/IIIa molecules and therefore cross-link receptors on adjacent activated platelets and ultimately lead to formation of platelet-rich thrombi. vWf, von Willebrand factor. (From Phillips and Scarborough,27 with permission.)

ial, occlusive red thrombi that produce complete blockage of a coronary artery in acute Q-wave MI. The arterial red thrombus is composed of red blood cells enmeshed within a fibrin network; however, it is built on a white thrombus core that creates regions of blood stasis conducive to fibrin deposition and provides platelet surfaces activating the coagulation cascade.1,4,5 Abrupt closure of a coronary artery, a common side effect of percutaneous transluminal coronary angioplasty (PTCA), has pathophysiologic features similar to that of other ACS. During the angioplasty procedure, mechanical injury to a focal atherosclerotic plaque can result in endothelial denudation and creation of fissures and tears. As a result, collagen and other thrombogenic connective tissue components are exposed, leading to platelet adhesion, activation, and thrombus formation.6,7 The realization that uncontrolled platelet aggregation could be responsible for thrombosis was appreciated as long ago as 1881, and the concept that thrombosis is the primary cause of ACS has been universally accepted in the last decade.1,5 Efforts to prevent ACS thus have logically focused on development of therapeutic interventions that block some component of normal hemostasis: platelet aggregation, coagulation, or both. Many distinct and sometimes redundant physiologic

pathways lead to platelet activation.1 The ultimate result of the action of these activation signals is platelet aggregation, mediated by cation-dependent attachment of divalent fibrinogen molecules to activated platelets. This cross-linking of platelets through fibrinogen bridges constitutes the final common pathway in platelet-mediated thrombus formation. The first evidence that the integrin glycoprotein (GP) IIb/IIIa is the receptor for this crucial fibrinogen-binding event came from studies of Glanzmann’s thrombasthenia, a rare, inherited, recessive bleeding disorder.8 Patients with Glanzmann’s thrombasthenia have a lifelong bleeding diathesis, defective clot retraction, and platelets that do not aggregate in response to any known physiologic activators. Patients with this disorder are characterized as having deletions, insertions, or point mutations in the genes specifying either subunit of the platelet receptor GP IIb/IIIa. Platelets from patients with Glanzmann’s thrombasthenia contain only 10% to 25% of the normal amount of α-granule fibrinogen and generally lack fibrinogen receptors on the cell surface.8,9 These observations pointed to a link between GP IIb/IIIa and fibrinogen binding. Other more recent studies reinforced this connection: monoclonal antibodies directed against GP IIb/IIIa have been found to block binding of

American Heart Journal Volume 138, Number 6, Part 1

fibrinogen to platelets and inhibit platelet aggregation; a fibrinogen derivative has been demonstrated to crosslink GP IIIa; and GP IIb/IIIa molecules reconstituted into lipid bilayers have been shown to bind fibrinogen.9 These and other studies have provided compelling evidence that fibrinogen binds to GP IIb/IIIa, an integrin found only on platelets and their precursors. Although GP IIb/IIIa is present on circulating platelets in large numbers (50,000 to 80,000 per platelet), it is unable to bind adhesive plasma proteins until the platelet is activated. On the surface of activated platelets, GP IIb/IIIa undergoes a conformational change that enables it to bind soluble fibrinogen, von Willebrand factor, fibronectin, and vitronectin. The binding to fibrinogen and von Willebrand factor leads to cross-linking of GP IIb/IIIa molecules on adjacent platelets and represents the final common pathway in platelet aggregation (Figure 1). The recognition of the role of GP IIb/IIIa in thrombus formation offered the opportunity to develop dramatically more effective antithrombotic therapies because agents inhibiting this receptor have the potential to prevent thrombosis regardless of the mechanisms leading to platelet activation. This review describes the development of eptifibatide, a small cyclic peptide that specifically blocks platelet receptor GP IIb/IIIa and shows promise as a safe and effective antithrombotic in the treatment of ACS.

Development of inhibitors of the platelet receptor GP IIb/IIIa The first GP IIb/IIIa inhibitor to be tested in clinical trials was a Fab fragment of an anti-GP IIb/IIIa murine monoclonal antibody, m7E3.10 Several small clinical trials demonstrated that m7E3 Fab can effectively block GP IIb/IIIa in human beings. It was evident from these studies that several potential safety concerns might be associated with this pharmacologic approach.11,12 Low titers of anti-m7E3 serum antibodies were detected in a number of healthy individuals and patients with stable coronary artery disease who received m7E3 Fab.13 m7E3 Fab also elicited an immune response in 36% of patients undergoing elective PTCA and in 52% of patients receiving thrombolytic therapy for acute MI.13,14 An increased incidence of thrombocytopenia has also been observed among patients treated with m7E3 Fab.12 In response to the antigenicity concerns with m7E3 Fab, a human-murine chimeric version of m7E3 Fab was developed.10 The resulting chimeric 7E3 IgG (c7E3) had binding properties identical to those of the original murine antibody, and a Fab fragment (c7E3 Fab; abciximab) of this chimeric IgG exhibited reduced antigenicity compared with its murine counterpart.10,14 Based on results from several smaller studies,10 a large phase III study, EPIC (Evaluation of c7E3 for the Prevention of

Scarborough 1095

Ischemic Complications) was initiated to evaluate the efficacy of c7E3 in high-risk patients undergoing PTCA. EPIC was a randomized, placebo-controlled, doubleblind, multicenter trial involving 2099 patients undergoing coronary angioplasty.15 Administration of abciximab as a bolus followed by infusion for up to 12 hours resulted in a statistically significant 35% reduction in the composite study end point, which consisted of the incidence of death, nonfatal MI, and the need for emergency revascularization at 30 days. On the other hand, the incidence of major bleeding episodes and the need for transfusions in patients who received abciximab were doubled compared with those of patients who received placebo.15 Clinical pharmacologic studies suggested that increased bleeding may be related to the slow rate at which abciximab dissociates from GP IIb/IIIa, leading to prolonged inhibition of hemostasis, particularly with the use of high doses of heparin.16,17 An increased incidence of thrombocytopenia was also observed as a result of treatment with abciximab.15 Even though the replacement of a part of the original mouse antibody with human sequences significantly reduced the antigenicity of m7E3 Fab, the resulting c7E3 Fab is a large protein with the potential for eliciting an immune response. In the EPIC trial, human antiabciximab antibodies were detected in 6.5% of patients receiving abciximab bolus and infusion, but these antibodies were not detected in patients receiving placebo.13,18 Limited data concerning the readministration of abciximab from a readministration registry suggest that readministered abciximab is safe with respect to bleeding effects, intracranial hemorrhages, efficacy, and anaphylaxis. However, an increase in severe thrombocytopenia was noted in patients retreated with abciximab (0.7% primary treatment, 2.2% with retreatment), which suggests that some caution should remain with any readministration of abciximab.19 The lack of specificity of abciximab for GP IIb/IIIa has also been debated. Abciximab binds to at least 2 other integrins, the broadly distributed vitronectin receptor and the leukocyte integrin Mac-1.20,21 The clinical significance of this integrin cross-reactivity remains a matter of speculation. Regardless of potential safety concerns, the success of abciximab in clinical studies is vitally important because it demonstrates that inhibition of the platelet receptor GP IIb/IIIa can be an effective treatment for patients undergoing PTCA. Because of the rather unique pharmacologic properties of abciximab, alternative means of blocking GP IIb/IIIa have also been extensively investigated. An alternate approach to blocking GP IIb/IIIa is to develop an agent that has a rapid rate of dissociation from platelet GP IIb/IIIa, such that its receptor-blocking activity is readily reversible at the termination of infusion, diminishing the risk of potential bleeding. It was also desirable to develop a GP IIb/IIIa inhibitor that did

1096 Scarborough

not elicit an immune response. Finally, the development of a GP IIb/IIIa inhibitor that would be highly specific for GP IIb/IIIa was also desirable. The close homology of the many integrin receptors poses a challenge in this respect, and the cross-reactivity of abciximab illustrates the difficulty of the task.

Synthesis of eptifibatide During the past decade, a wide range of snake viper venoms have been shown to contain a family of compounds called disintegrins, small proteins that act as powerful antithrombotics and strongly inhibit platelet aggregation.22,23 Almost all these small proteins contain the amino acid sequence RGD (Arg-Gly-Asp), which is believed to play a role in the binding of fibrinogen to GP IIb/IIIa. However, RGD-containing venom disintegrins also bind to a number of other integrins, such as α5β1 (fibronectin receptor) and αvβ3 (vitronectin receptor) and therefore lack the specificity for GP IIb/IIIa.24 In an effort to find a disintegrin specific for GP IIb/IIIa, we screened 62 different snake venoms, 52 of which exhibited some type of integrin-binding activity. Only barbourin, a 73-amino acid disintegrin from the venom of the southeastern pigmy rattlesnake Sistrurus barbouri, selectively inhibited fibrinogen binding to GP IIb/IIIa and did not interfere with binding of RGDcontaining ligands to other integrins, including α5β1 and αvβ3.24,25 Barbourin was sequenced and found to be closely homologous to other disintegrins found in viper venoms, but it did not contain the RGD sequence. The binding of barbourin to GP IIb/IIIa was instead attributable to the amino acid sequence KGD (Lys-GlyAsp). The Lys substitution appears to be the sole structural feature imparting GP IIb/IIIa specificity, because venom peptide analogs with Lys substitutions, including a synthetic, truncated form of barbourin containing the KGD motif, are all specific for GP IIb/IIIa. However, the activity of both barbourin and other disintegrins strongly depends on their tertiary structure. Native disintegrins have numerous disulfide bridges and are approximately a thousand times more potent as inhibitors than are linear RGD-containing peptides. Also, although linear KGD-containing peptides are about 6-fold less active as antagonists of GP IIb/IIIa than are corresponding linear RGD-containing peptides, barbourin and Lys-substituted venom peptides appear to be just as potent as the RGD disintegrins. This suggests that the appropriate conformational display of the KGD sequence permits high-affinity binding while maintaining receptor selectivity.24-26 Barbourin is a fairly large peptide and is unsuitable for clinical application because of its rapid elimination and potential antigenicity.25 However, it provided a useful template for development of nonimmunogenic smaller peptides resistant to proteolytic degradation yet would have a rapid elimination from the circulation. A

American Heart Journal December 1999

detailed dissection of the structural components of barbourin established that its specificity is indeed determined by the KGD motif. The affinity for GP IIb/IIIa, however, is profoundly influenced by the amino acid residues immediately adjacent to the KGD sequence and the size of the peptide ring created through the formation of a disulfide bond. This information was used in the design of synthetic peptides with potential for clinical use. The peptides were designed to be small, containing 10 amino acids or less, to minimize their immunogenic potential. In addition, the peptides had to be cyclic to preserve the high-affinity tertiary structure of the receptor-binding sequence and to ensure resistance to in vivo proteolytic degradation. With these limitations in mind, peptides that are both potent and highly specific antagonists of GP IIb/IIIa and that closely mimic the properties of the much larger, 73amino acid barbourin molecule have been designed.26-28 The efforts to create synthetic analogues of barbourin eventually resulted in eptifibatide (Integrilin, COR Therapeutics, South San Francisco, Calif), a heptapeptide cyclized by a disulfide bridge. Eptifibatide is a potent and specific inhibitor of GP IIb/IIIa and contains a modified KGD peptide recognition sequence.28 Eptifibatide, like barbourin, binds with high affinity to GP IIb/IIIa and is the only barbourin-derived cyclic compound to be developed for clinical use so far.

Pharmacology of eptifibatide Eptifibatide, like other antagonists of platelet receptor GP IIb/IIIa, functions by blocking the binding of the adhesive proteins fibrinogen and von Willebrand factor to GP IIb/IIIa on the surface of activated platelets. It is a potent antithrombotic because the binding of these proteins to GP IIb/IIIa is the event that precipitates platelet aggregation and subsequent arterial thrombus formation. As a critical prelude to clinical trials, studies were undertaken to determine the relation between plasma levels of eptifibatide and the degree of inhibition of platelet aggregation and thrombosis. From these investigations, ranges of potentially effective eptifibatide dosages have been derived and applied in the design of both small- and large-scale clinical trials. Overall, eptifibatide rapidly inhibits ex vivo platelet aggregation in blood samples, with an IC50 in the 100 to 600 nmol/L range, depending on the extracellular concentration of ionized calcium; at a low concentration of 40 to 50 µmol/L, the IC50 is approximately 100 nmol/L, whereas at physiologic concentration of approximately 1.1 µmol/L, IC50 is closer to 500 nmol/L.29 Eptifibatide is a potent and specific inhibitor of GP IIb/IIIa, and because GP IIb/IIIa is confined to platelets and their precursors, the pharmacologic action of eptifibatide is essentially confined to those cells, thus minimizing the potential for side effects and making the drug more attractive for clinical use. The effectiveness

American Heart Journal Volume 138, Number 6, Part 1

of eptifibatide as an inhibitor of GP IIb/IIIa has been confirmed in a variety of preclinical and clinical studies. These studies showed a rapid inhibition of platelet aggregation within 15 minutes of eptifibatide treatment, and at appropriate dosages this inhibition was sustained for the duration of the drug infusion.30 Eptifibatide is also characterized by a short pharmacokinetic and pharmacodynamic half-life (1 to 2 hours) after the treatment is terminated. The rate of eptifibatide elimination from circulation is rapid (1 to 2 hours).28 Normal platelet function is similarly restored within 2 to 4 hours after treatment termination, irrespective of the dosing regimen.31 A short half-life may confer an advantage for clinical use because it reduces the risk of serious bleeding and decreases recovery time after treatment, thus potentially shortening the duration of hospital stays. Thus far in clinical trials, eptifibatide has not been found to be immunogenic or to induce thrombocytopenia. In the phase III IMPACT II trial, no antibodies to eptifibatide were detected at 30 days after treatment, a result anticipated because of the small size (7 amino acids) of eptifibatide. Its cyclic structure, as noted, renders eptifibatide resistant to degradation by plasma proteases and increases its plasma stability. These features, most notably rapid reversibility, specificity, and lack of immunogenicity, have provided the impetus for pharmacologic and therapeutic evaluations of eptifibatide that sought to validate its clinical potential as an alternative to other antiplatelet agents in the management of ACS.

Antiplatelet activity of eptifibatide Determination of an ex vivo surrogate to predict in vivo activity of eptifibatide has relied primarily on its effects in an ex vivo platelet aggregation assay. In all early clinical studies, this assay was routinely performed in blood anticoagulated with citrate. Citrate prevents in vitro clotting of blood samples by reducing the ionized calcium concentration from the 1.1 mmol/L normally found in circulating blood to 40 to 50 µmol/L. Therefore ionized calcium concentration has been significantly reduced in most pharmacodynamic measurements of platelet aggregation inhibition. As described, it was later understood that this property of citrate anticoagulation distorted the apparent relation between eptifibatide dose and the degree of inhibition of platelet aggregation.29 GP IIb/IIIa molecules can bind ionized calcium and other divalent cations through 5 potential ionized calcium binding sites, which are all occupied in buffers containing physiologic levels of ionized calcium (1 mmol/L). Divalent cation binding to GP IIb/IIIa is required for binding of fibrinogen and therefore for platelet aggregation. If the platelets are suspended in a medium containing only 20 to 100 µmol/L ionized calcium, the ionized calcium dissociates from GP IIb/IIIa

Scarborough 1097

on the platelet surface. At ionized calcium concentrations below 1 µmol/L at physiologic temperature, GP IIb dissociates from GP IIIa within the plane of the plasma membrane, and both subunits lose their structure. If the concentration of ionized calcium is only moderately reduced, the affinity of fibrinogen for GP IIb/IIIa on activated platelets is lowered, but sufficient binding persists to allow platelet aggregation.28,29 Reduced ionized calcium also simultaneously increases the binding of eptifibatide, possibly because eptifibatide and ionized calcium occupy overlapping sites on GP IIb/IIIa. In a direct evaluation of the role of ionized calcium concentration, the ability of eptifibatide to inhibit fibrinogen binding to purified GP IIb/IIIa was compared at different calcium concentrations.29 At 1 mmol/L ionized calcium, eptifibatide inhibited fibrinogen binding to receptor with an IC50 of 63 nmol/L. At 50 µmol/L ionized calcium, eptifibatide was a more potent inhibitor, with an IC50 of 8.7 nmol/L. Thus the inhibitory effect of eptifibatide is markedly enhanced in buffers with reduced concentrations of ionized calcium. This enhancement of eptifibatide activity is therefore observed in blood samples anticoagulated with citrate, which chelates ionized calcium and can lead to overestimates of the in vivo pharmacologic effects of the drug. More accurate estimates of in vivo activity of eptifibatide may be obtained if blood is collected in buffers containing PPACK (Phe-Pro-Arg chloromethyl ketone), an anticoagulant that does not chelate ionized calcium and alter the physiologic ionized calcium concentration. To evaluate this hypothesis, ex vivo inhibition of adenosine diphosphate (ADP)-activated platelet aggregation by eptifibatide was assessed in blood anticoagulated with citrate or PPACK (Figure 2).29 At 500 nmol/L eptifibatide, platelet aggregation was completely inhibited in blood anticoagulated with citrate, but the inhibition was only 15% to 30% in PPACK-anticoagulated blood. In fact, the concentration of eptifibatide needed for 50% inhibition of ADP-induced platelet aggregation was 4-fold greater for plasma anticoagulated with PPACK than with citrate. If thrombin receptor agonist peptide (TRAP) is used for platelet activation in place of ADP, a still higher concentration of eptifibatide is required to achieve the 50% inhibition of aggregation (Figure 2), probably because TRAP induces the exposure of additional GP IIb/IIIa molecules on the platelet surface. However, the physiologic relevance of platelet activation by TRAP is not clear, and ADP-induced platelet activation continues to be routinely used for estimating in vivo antithrombotic activity of GP IIb/IIIa inhibitors. In conclusion, a decrease in ionized calcium concentration from physiologic (1 mmol/L) to that in citrate buffer (50 µmol/L) enhances GP IIb/IIIa binding of eptifibatide and thereby increases its apparent inhibitory

American Heart Journal December 1999

1098 Scarborough

Figure 2

Inhibition of platelet aggregation by eptifibatide. Platelet-rich plasma samples were anticoagulated with citrate or PPACK-incubated with various concentrations of eptifibatide. Platelet aggregation was determined in response to 20 µmol/L ADP or 5 µmol/L TRAP. Steady-state plasma concentrations of eptifibatide achieved with dosing regimens used in IMPACT II study are indicated. (From Phillips and Scarborough,27 with permission.)

activity. In this way, use of citrate-anticoagulated blood grossly overestimates the platelet aggregation inhibitory activity of eptifibatide in vivo for a particular dosage regimen. This finding had important implications for the interpretation of results from early clinical studies, because dosing of eptifibatide in these studies was based on measurements of platelet aggregation in blood anticoagulated with citrate.

Pharmacodynamics of eptifibatide in animal models Several preclinical experiments to test the efficacy and safety of eptifibatide in nonhuman species produced promising results. An animal study with a canine model to investigate the effects of eptifibatide on fibrinogen-triggered platelet adhesion and postoperative bleeding related to cardiopulmonary bypass found a rapidly reversible antihemostatic effect of eptifibatide, which was associated with seemingly paradoxic minimization of postoperative bleeding. The reduction of postoperative bleeding probably stems from inhibition of platelet adhesion to fibrinogen bound to foreign surfaces such as tubing and oxygenator during the procedure, ultimately resulting in the preservation of hemostatically competent platelets dur-

ing the bypass procedure.31 In another canine study, the infusion of 4 µg/kg/min of eptifibatide effectively inhibited ex vivo platelet aggregation in citrate-anticoagulated blood with no evidence of prolonged buccal bleeding time.28 In a study performed in a baboon model of thrombosis, complete inhibition of ADP-induced ex vivo platelet aggregation in citrate-anticoagulated blood was obtained within 25 minutes of the initiation of eptifibatide infusions of 5 µg/kg/min and 10 µg/kg/min. Spontaneous bleeding was not observed with either dose, and the increase in simplate bleeding time was modest. Platelet aggregation returned to normal levels within 15 to 30 minutes of the cessation of infusion, demonstrating the rapid reversibility of GP IIb/IIIa inhibition by eptifibatide. Bleeding times normalized more rapidly than platelet aggregation and were within the normal range if their measurements were initiated on cessation of the eptifibatide infusion.28

Clinical pharmacokinetics and pharmacodynamics of eptifibatide Studies in healthy volunteers A potent and rapidly reversible inhibition of GP IIb/IIIa by eptifibatide was demonstrated in healthy vol-

American Heart Journal Volume 138, Number 6, Part 1

unteers in a variety of clinical settings. For infusions ranging from 0.2 to 1.5 µg/kg/min, plasma concentrations of eptifibatide were proportional to the dose and rate of eptifibatide administration. In these studies, the plasma elimination half-life of eptifibatide was 50 to 60 minutes. In a phase I investigation of 63 healthy volunteers, ADP-stimulated platelet aggregation was completely inhibited (in citrate-anticoagulated blood) by a constant infusion of eptifibatide (1.0 to 1.5 µg/kg/min), an effect that was reversed within 2 to 4 hours of termination of the infusion. Simplate bleeding times were slightly prolonged during the infusion, but they returned to normal within 30 minutes after infusion. There were no clinical bleeding events reported for the phase I investigation.32

Studies in patients with unstable angina Two early dose-ranging phase II clinical trials were undertaken to evaluate the efficacy of eptifibatide in patients with unstable angina. The first was designed to evaluate the pharmacodynamics of eptifibatide in 61 patients with unstable angina. Three doses were studied, ranging from 120 µg/kg bolus and 1.0 µg/kg/min infusion to 150 µg/kg bolus and 1.25 µg/kg/min infusion. Overall, ADP-stimulated ex vivo platelet aggregation in citrate-anticoagulated blood was reduced by 70% to 80% from baseline levels.33 The second study, involving 227 patients, was a prospective, double-blind, multicenter trial designed to compare the number and duration of ischemic episodes in patients with unstable angina randomly assigned to receive eptifibatide and heparin or aspirin and heparin over a 24- to 72-hour period. Ischemia during the first 24 hours of study drug infusion was detected by Holter monitoring and occurred in 21% of the patients randomly assigned to aspirin, 20% of those receiving the lower dose of eptifibatide (45 µg/kg bolus followed by 0.5 µg/kg/min) and 10% of the patients receiving the higher dose of eptifibatide (90 µg/kg bolus and 1.0 µg/kg/min infusion).34 When ischemia determined by electrocardiogram did occur, it was of shorter duration with the higher dose of eptifibatide than with aspirin. Termination of eptifibatide infusion was not associated with rebound ischemia.34 Both dosing regimens of eptifibatide produced a rapid, dose-dependent inhibition of ex vivo platelet aggregation within an hour of the start of the infusion. At 1 hour and at 4 hours of infusion, as well as at the end of the drug administration, the inhibition of platelet aggregation obtained with either dose of eptifibatide was significantly greater than that obtained with aspirin. This eptifibatide-mediated inhibition was rapidly reversible, and the platelet aggregation returned to 60% of baseline levels within 4 hours of termination of the infusion. Eptifibatide was well tolerated in this study. There were few clinical

Scarborough 1099

events and no significant differences in serious bleeding among the treatment arms.34 These studies indicated that high-dose eptifibatide is a potent inhibitor of platelet aggregation that reduces the frequency and duration of electrocardiographically determined ischemia in patients with unstable angina to a greater degree than does standard antiplatelet therapy.

Study in patients with acute MI Successful reperfusion of an infarct-related artery is the main predictor of survival after acute MI. A placebo-controlled, dose-ranging clinical trial involving 180 patients with acute MI assessed the effects of eptifibatide on reperfusion when given as an adjunct to thrombolysis with alteplase (a tissue plasminogen activator).35 Increasing dosage regimens of eptifibatide, ranging from 36 µg/kg bolus and 0.2 µg/kg/min infusion to 180 µg/kg bolus and 0.75 µg/kg/min infusion, were begun within 30 minutes of initiation of alteplase therapy. Platelet inhibition, measured by ex vivo aggregation assays in citrate-anticoagulated blood, was maximal (70% to 80% for the higher doses) at 2 hours. For the highest dose tested, the platelet aggregation remained suppressed by 70% at 24 hours after the start of eptifibatide infusion. Paralleling this platelet inhibition, a significantly greater degree of culprit artery reperfusion was achieved in patients given the highest dose of eptifibatide compared with that achieved in placebo-treated patients (66% vs 39% for placebo; P = .006). Patients who received the highest dose of eptifibatide also had a shorter median time to ST-segment recovery (65 vs 116 minutes for placebo; P = .05).35 The results indicate that improved coronary artery reperfusion in patients with MI can be achieved when thrombolytics are combined with GP IIb/IIIa inhibitors.

Studies in patients undergoing elective PTCA A series of small dose-finding studies were carried out in which eptifibatide was given to patients scheduled to undergo elective PTCA. The results of these studies were instrumental in designing a subsequent large phase III trial. In the phase II, multicenter IMPACT (Integrilin to Minimize Platelet Aggregation and Coronary Thrombosis) trial, 150 patients were randomly assigned to receive either eptifibatide (90 µg/kg bolus followed by 1.0 µg/kg/min infusion) or placebo before and for 4 or 12 hours after the elective PTCA.36 The eptifibatide regimen inhibited ex vivo, ADP-activated platelet aggregation (in citrate-anticoagulated blood) by 86%, a level that was maintained throughout the infusion period. This study represented the first clinical investigation of eptifibatide during routine, elective and low- and high-risk coronary intervention and supported the potential efficacy of eptifibatide in routine coronary interventions.

1100 Scarborough

A second phase II study involving 73 patients undergoing elective PTCA, the IMPACT High/Low Trial, was designed to establish more firmly the safety of several eptifibatide dosing regimens and to determine the dose relation of the pharmacologic effect, as measured by ex vivo platelet aggregation assays in citrate-anticoagulated blood.30 Patients scheduled for routine angioplasty received placebo or 1 of 4 dosing regimens of eptifibatide, administered as a bolus followed by an 18- to 24-hour infusion: (1) 180 µg/kg bolus plus 1.0 µg/kg/min infusion, (2) 135 µg/kg bolus plus 0.5 µg/kg/min infusion, (3) 90 µg/kg bolus plus 0.75 µg/kg/min infusion, (4) 135 µg/kg bolus plus 0.75 µg/kg/min infusion. In more than 75% of patients receiving the bolus of 135 µg/kg, a rapid (15 min after bolus) inhibition of ADP-stimulated ex vivo platelet aggregation (>80%) was observed.30 Even though the IMPACT High/Low Trial was not designed to test the effectiveness of eptifibatide in preventing ischemia in patients undergoing coronary angioplasty, consistently fewer ischemic complications were observed in the patients treated with eptifibatide than in the placebo group. The results of this study supported the concept that potent GP IIb/IIIa inhibition in combination with heparin and aspirin administered at the time of coronary intervention is associated with fewer periprocedural ischemic complications than treatment with heparin and aspirin alone. The development of eptifibatide was undertaken with the intention to provide a potent antiplatelet agent with a favorable safety profile. The preliminary studies described provided a strong argument in favor of the concept that eptifibatide should offer an attractive safety profile with low risk of bleeding, which is particularly important if emergency surgery is required after the initial treatment. Eptifibatide did not demonstrate antigenic potential, which should permit repeated administration without serious risk of thrombocytopenia or development of an antibody response. In the IMPACT and IMPACT High/Low trials, sample sizes were too small to demonstrate a statistically significant effect on clinical outcomes, but there clearly was a dose-dependent antithrombotic effect and a trend toward a reduction in ischemic events, including death, MI, and repeat revascularization. The IMPACT High/Low Trial also demonstrated a rapid and rapidly reversible inhibitory effect of eptifibatide on GP IIb/IIIa and provided guidelines for dosing regimens in the subsequent phase III IMPACT II trial.

Large clinical trials of eptifibatide The IMPACT-II trial IMPACT II was a randomized, double-blind, placebocontrolled phase III clinical trial designed to investigate the efficacy of eptifibatide in reducing ischemic compli-

American Heart Journal December 1999

cations of percutaneous coronary revascularization in a broad range of patients.37 The study was performed in 84 centers in the United States between November 1993 and November 1994 and involved 4010 patients scheduled for either elective, urgent, or emergency coronary intervention (balloon angioplasty, directional coronary atherectomy, high-speed rotational atherectomy, or excimer laser ablation). Elective stent implantation was not included because it had not been approved at the time. Patients excluded from the study included those who had a history of bleeding diathesis, severe hypertension, major surgery within the previous 6 weeks, previous stroke or other central nervous system disease, gastrointestinal or genitourinary bleeding within the previous 30 days, or other major comorbid illness. Pregnant women were also excluded. Patients were randomly assigned to 1 of 3 treatment regimens derived from the results of the IMPACT High/Low Trial: (1) placebo bolus and infusion, (2) 135 µg/kg bolus of eptifibatide followed by 0.5 µg/kg/min infusion for 20 to 24 hours (135/0.5), or (3) 135 µg/kg bolus plus 0.75 µg/kg/min infusion for 20 to 24 hours (135/0.75). The bolus dose selected was shown to provide robust platelet inhibition during and immediately after angioplasty, and the 2 infusion regimens were selected to identify an optimal dose that would maintain clinical benefit without compromising safety. All patients received aspirin (325 mg) before the procedure. After vascular access was established, a bolus of 100 U/kg of heparin was given, and infusion of the study drug was initiated. Activated clotting time was obtained 5 minutes after the study drug bolus was administered, and additional heparin was given as needed to maintain an activated clotting time of 300 to 350 seconds throughout the angioplasty procedure. Both eptifibatide groups had prolonged activated clotting times compared with those of the placebo group. Stent implantation was not permitted unless required to treat abrupt vessel closure. Blood samples were obtained 6, 12, and 24 hours after the study drug bolus administration for assessment of platelet counts and creatine kinase MB levels. The mean plasma concentrations of eptifibatide, obtained from 77% of the patients at the termination of drug infusion, were 270 ng/mL and 388 ng/mL for the 135/0.5 and 135/0.75 dosage groups, respectively. Serum samples were obtained from 390 patients, representing all trial centers, 30 days after the procedure. None of the samples contained anti-eptifibatide antibodies, confirming the lack of antigenicity of the peptide.37 The primary end point of the trial was the composite incidence of death, MI, and urgent revascularization within 30 days of the procedure. The results showed a clear early benefit of eptifibatide administration across a spectrum of ischemic complications; at 24 hours both eptifibatide regimens significantly reduced the composite incidence of death, MI, and need for urgent revascu-

American Heart Journal Volume 138, Number 6, Part 1

Scarborough 1101

Table I. Efficacy of eptifibatide in the IMPACT II trial: Incidence of composite end point in treated patients at 24 hours, 48 hours, and 30 days after randomization

24 h 48 h 30 d

Placebo (n = 1285)

Eptifibatide 135/0.5 (n = 1300)

Eptifibatide 135/0.75 (n = 1286)

9.6% 10.2% 11.6%

6.6%* 7.6%* 9.1%*

6.9%* 7.9%* 10.0%

*P < .05 vs placebo.

larization compared with that for the placebo group, and the effect was maintained for at least 72 hours. The incidence of abrupt vessel closure was also significantly reduced in patients treated with either eptifibatide dosage regimen at 24 hours: 2.8% and 3.5% for the 135/0.5 and 135/0.75 dosage groups, respectively, vs 5.1% for the placebo-treated patients.37,38 Patients who received eptifibatide therapy had a lower incidence of the composite end point at 30 days compared with those treated with placebo, suggesting a lasting benefit of the effects achieved immediately after angioplasty (Table I). Among patients who received eptifibatide treatment, the reductions in the composite end point at 30 days were 22% (P = .035) and 14% (P = 0.178) for the 135/0.5 and 135/0.75 doses, respectively. Reductions were demonstrated for all individual end point events as well, emphasizing the trend, although these reductions did not reach statistical significance.37,38 For the all randomized population, the reductions in the composite end point at 30 days were 21% (P = .0630) and 14% (P = .22) for the 135/0.5 and 135/0.75 doses, respectively. With this analysis, neither dose achieved the significance level (P < .035) for efficacy. The data from patients enrolled in the IMPACT II trial were analyzed again at 6 months. These results were similar to those obtained at 30 days, indicating that the trends observed early on had persisted for 6 months (Figure 3). Although not reaching statistical significance, the most pronounced reductions were seen in the composite rate of death and MI (10.9% and 13.9% reduction for the 135/0.5 and 135/0.75 doses, respectively).38 The IMPACT II study clearly demonstrated the excellent safety profile of eptifibatide. Major bleeding events, severe bleeding, red blood cell transfusions, and other adverse effects were similar in all treatment groups, whereas the proportion of minor bleeding events was slightly higher in patients given eptifibatide. No differences among the treatment groups were observed with respect to the minimum platelet count, frequency of thrombocytopenia, net change in platelet count, or need for platelet transfusions.35,38 In summary, the IMPACT II trial demonstrated clinical benefits for eptifibatide in all patients scheduled to undergo coronary angioplasty. The majority of ischemic

complications of PTCA (60% to 70%) occur within 6 hours of the procedure, and the substantial, statistically significant early effects of eptifibatide treatment are in accord with the importance of platelet aggregation in the development of these complications. This beneficial effect was not accompanied by an increased rate of bleeding complications, supporting the acceptability of eptifibatide as a clinical alternative to other antiplatelet agents.

Pharmacologic findings after IMPACT II After the IMPACT II trial was completed, significant new findings were made regarding the dosing regimens of eptifibatide in this study. Plasma concentrations of eptifibatide during infusion were at the levels expected to inhibit ADP-stimulated platelet aggregation in citrate-anticoagulated blood by 70% to 100%. However, as previously discussed, the use of citrate for anticoagulation in platelet aggregation assays leads to a decrease in ionized calcium concentration, and the in vivo antiplatelet effects of eptifibatide are therefore overestimated. In PPACK-anticoagulated blood, which is more representative of in vivo physiologic conditions, the degree of inhibition of ADP-induced platelet aggregation by the infusion doses of eptifibatide used in IMPACT II would be significantly lower (35% to 50%).29 It is now apparent that the dosing regimens used in the IMPACT II trial were at the low end of the dose-response curve (Figure 2). The significant early benefits observed in this trial were probably attributable to the bolus dose, which achieved transient plasma levels of eptifibatide >600 nmol/L, sufficient to establish a rapid and robust inhibition of platelet aggregation. The similar results obtained with the 2 infusion doses of eptifibatide can probably be explained by the potent effect of the bolus dose, which was the same in the 2 groups. In addition, considerable standard deviations in mean plasma levels of the drug were observed during the infusion period, which resulted in significant overlap between the 2 experimental arms. Although the infusion doses of eptifibatide used in the IMPACT II trial were at the low end of the dose-response curve, this trial nevertheless demonstrated a sustained benefit of eptifibatide during and after coronary angioplasty.

1102 Scarborough

American Heart Journal December 1999

Figure 3

Kaplan-Meier curve of frequency of death and MI in treated patients in IMPACT II. (From Tcheng,39 with permission.)

The PURSUIT trial The findings of the IMPACT II trial, as well as the realization of caveats inherent in performing platelet aggregation assays in blood anticoagulated with citrate, suggested that higher doses of eptifibatide may lead to even greater beneficial effects of this agent. The efficacy of increased doses was tested in the PURSUIT (Platelet GP IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy) trial, a worldwide phase III evaluation of eptifibatide in the setting of unstable angina and non-Q-wave MI involving 10,948 patients from more than 700 hospitals in 28 countries.39,40 The PURSUIT trial is the largest clinical evaluation of a GP IIb/IIIa inhibitor to date as well as the largest trial of unstable angina and non-Q-wave MI. Patients eligible for enrollment were those who had prolonged (≥10 minutes) acute chest pain of suspected ischemic origin within the previous 24 hours. Also, patients must have had either specific electrocardiographic changes (transient <30-minute ST-segment elevation >1.0 mm, transient or persistent ST-segment depression >0.5 mm, or T-wave inversions >1.0 mm) within 12 hours of ischemia or positive creatine kinase assay results. Excluded patients were those who had had major bleeding episodes within the previous 30 days, patients with a history of bleeding diathesis, and those who had received thrombolytic agents within 24 hours of enrollment.39,40 The specific dosing regimen chosen for the PURSUIT trial was based on an ex vivo platelet aggregation assay that used PPACK as the anticoagulant, maintaining the

ionized calcium concentration close to physiologic levels. This assay provided a more stringent test of the GP IIb/IIIa inhibitory activity of eptifibatide than assays in the IMPACT II trial, which used citrate as the anticoagulant. Initially patients were randomly assigned to receive placebo or 1 of the 2 eptifibatide dosing regimens: a bolus of 180 µg/kg and infusion of 1.3 µg/kg/min or a bolus of 180 µg/kg and infusion of 2.0 µg/kg/min. The infused drug was administered for up to 72 hours or for up to 96 hours if PTCA was performed on day 3 of therapy. This infusion rate, approximately 3- to 4-fold higher than that used in IMPACT II, was estimated to provide a minimum of 80% of inhibition of ex vivo, ADP-activated platelet aggregation in PPACK-anticoagulated blood. All patients received aspirin. Heparin, other adjunct therapies, and interventional procedures were used at the discretion of the investigator. The study design specified that one of the treatment arms be discontinued after an interim safety analysis. At the interim analysis, the Data and Safety Monitoring Committee determined that the higher dose of eptifibatide demonstrated excellent safety and recommended discontinuation of the lower-dose eptifibatide arm. From that point, patients were randomly assigned to receive either placebo or the eptifibatide bolus of 180 µg/kg followed by 2.0 µg/kg/min infusion.39,40 The primary efficacy end point of the PURSUIT trial was death or MI at 30 days, and the calculated sample size was based on an estimated 8.5% incidence of death or MI at this time point in the placebo arm. Given a

American Heart Journal Volume 138, Number 6, Part 1

large sample size, a reduction of 14% in death or MI would be statistically significant. Secondary end points included incidence of death or (re)infarction at 96 hours, 7 days, and 6 months. The rates of death or MI in patients receiving PTCA and those treated medically, as well as the cost of treatment and the quality of patient life, were also subjects of secondary analyses. At 30 days, eptifibatide treatment resulted in a statistically significant reduction in the incidence of death or MI from 15.7% to 14.2% (P = .04), which corresponds to 15 fewer events per 1000 patients treated.40 This 1.5% absolute reduction in events was documented early on within 96 hours of treatment (9.1% for placebo vs 7.6% for eptifibatide, P = .011) and sustained for the duration of the study. The reduction in death or MI was seen in both patients treated medically and those undergoing PTCA. Among patients who underwent PTCA within 72 hours of randomization, those treated with eptifibatide had reduced rates of death or MI both before the intervention was performed (1.5% vs 4.2% in the placebo group) and 30 days after the intervention was performed (11.8% vs 16.8% in the placebo group). Subanalyses revealed large regional variations in study results; the effect of eptifibatide treatment was most robust in North America, somewhat less pronounced in Western Europe, and did not occur in Eastern Europe and Latin America. The reasons for this geographic variability of study results are not clear but may include differences in baseline characteristics of the patients as well as in treatment practices. In addition, stringent adjudication criteria for MI were used by the Clinical Endpoints Committee, and any differences in accuracy of diagnostic procedures (measurement of cardiac enzyme levels) would be reflected in the recordings of end point events. On that note, it should be mentioned that the analysis of events as assessed by the investigators demonstrated much lower variability and consistent benefit of eptifibatide across all geographic regions. The greatest benefit of eptifibatide therapy was recorded among patients in the United States, who represented 37% of the overall study population. In the US centers, eptifibatide treatment led to a 3.5% absolute reduction in end point events at 30 days, from 15.4% to 11.9% (P ≤ .003).40,41 Although interventional procedures were used more frequently in the United States than in the other study regions, this difference alone cannot account for the differences in the study results. In the US centers, treatment with eptifibatide was effective in reducing end point events in both patients undergoing interventional procedures (11.4% vs 16.8% for placebo, P < .05) and those receiving only medical therapy (12.1% vs 14.9% for placebo, P < .05).40 The incidence of major bleeding was 9.3% for placebo-treated patients and 10.8% for patients treated with eptifibatide. Treatment with eptifibatide did not lead to an increase in strokes, and there was no addi-

Scarborough 1103

tional risk incurred by patients undergoing coronary artery bypass graft. The incidence of thrombocytopenia in the 2 treatment arms was similar (6.8% in the eptifibatide group and 6.7% in the placebo group; P = .864).

Summary There are several physiologic cascades that ultimately lead to the aggregation of platelets and thrombosis, but all converge on the platelet receptor GP IIb/IIIa. The interaction of activated GP IIb/IIIa with its ligands represents the final common pathway for platelet aggregation. For this reason, a GP IIb/IIIa blockade has promise as an almost ideal antithrombotic therapy, overcoming the limitations of traditional antithrombotic therapies. Because GP IIb/IIIa is found only on platelets and their precursors, a specific GP IIb/IIIa inhibitor is not expected to interfere with physiologic processes mediated by other cell types. Therefore the incidence of adverse effects of treatment should be minimal.

References 1. Coller BS. Platelets in cardiovascular thrombosis and thrombolysis. In: Fozzard HA, Haber E, Jennings RB, Katz AM, Morgan HE, editors. The heart and cardiovascular system: scientific foundations. 2nd ed. New York: Raven Press; 1992. p. 219-73. 2. Coute S, Leung L. Novel antithrombotic therapeutics targeted against platelet glycoprotein IIb/IIIa. Ann Rev Med 1995;46:257-65. 3. Gotoh K, Minamino T, Katoh O, Hamano K, Fukui S, Hori M, et al. The role of intracoronary thrombus in unstable angina: angiographic assessment and thrombolytic therapy during ongoing anginal attacts. Circulation 1988;77:526-34. 4. Davies MJ. A macro and micro view of coronary vascular insult in ischemic heart disease. Circulation 1990;82(suppl II):II-38-II-46. 5. DeWood MA, Spores J, Notske R, Mouser LT, Burroughs R, Golden MS, et al. Prevalence of total coronary occlusion during the early hours of transmural myocardial infarction. N Engl J Med 1980; 303:897-902. 6. Landau C, Lange RA, Hillis LD. Percutaneous transluminal coronary angioplasty. N Engl J Med 1994;330:981-93. 7. Wilson JM, Silberman H, Ferguson JJ. Abrupt vascular closure. J Invas Cardiol 1994;6:306-13. 8. Coller BS, Seligsohn U, Peretz H, Newman PJ. Glanzmann thrombasthenia: new insights from an historical perspective. Semin Hematol 1994;31:301-11. 9. Phillips DR, Charo IF, Parise LV, Fitzgerald LA. The platelet membrane glycoprotein IIb/IIIa complex. Blood 1988;71:831-43. 10. Jordan RE, Mascelli MA, Nakada MT, Weisman HF. Pharmacology and clinical development of abciximab (c7E3 Fab, ReoPro): a monoclonal antibody inhibitor of GPIIb/IIIa and αvβ3. In: Sasahara AA, Loscalzo J, editors. New therapeutic agents in thrombosis and thrombolysis. New York: Marcel Dekker; 1997. p. 291-313. 11. Ellis SG, Tcheng JE, Navetta FI, Muller DWM, Weisman HF, Smith C, et al. Safety and antiplatelet effect of murine monoclonal antibody 7E3 Fab directed against platelet glycoprotein IIb/IIIa in patients undergoing elective coronary angioplasty. Coron Art Dis 1993;4:167-75. 12. Kleiman NS, Ohman EM, Califf RM, George BS, Kereiakes D, Aguirre FV, et al. Profound inhibition of platelet aggregation with monoclonal antibody 7E3 Fab after thrombolytic therapy. Results of

1104 Scarborough

13.

14. 15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

the Thrombolysis and Angioplasty in Myocardial Infarction (TAMI) 8 pilot study. J Am Coll Cardiol 1993;22:381-9. Knight DM, Wagner C, Jordan R, McAleer MF, DeRita R, Fass DN, et al. The immunogenicity of the 7E3 murine monoclonal Fab antibody fragment variable region is dramatically reduced in humans by substitution of human for murine constant regions. Mol Immunol 1995;32:1271-81. Coller BS. Blockade of platelet GPIIb/IIIa receptors as an antithrombotic strategy. Circulation 1995;92:2373-80. The EPIC Investigators. Use of a monoclonal antibody directed against the platelet glycoprotein IIb/IIIa receptor in high-risk coronary angioplasty. N Engl J Med 1994;330:956-61. Tcheng JE, Ellis SG, George BS, Kereiakes DJ, Kleiman NS, Talley JD, et al. Pharmacodynamics of chimeric glycoprotein IIb/IIIa integrin antiplatelet antibody Fab 7E3 in high-risk coronary angioplasty. Circulation 1994;90:1757-64. Konstantopoulos K, Kamat SG, Schafer AI, Baez EI, Jordan R, Kleiman NS, et al. Shear-induced platelet aggregation is inhibited by in vivo infusion of an anti-glycoprotein IIb/IIIa antibody fragment, c7E3 Fab, in patients undergoing coronary angioplasty. Circulation 1995;91:1427-31. Faulds D, Sorkin EM. Abciximab (c7E3 Fab). A review of its pharmacology and therapeutic potential in ischaemic heart disease. Drugs 1994;48:583-98. Tcheng JE, Kereiakes DJ, Braden GA, Lincoff AM, Mascelli MA, Langrall MA, et al. Safety of abciximab retreatment: final clinical report of the ReoPro readministration registry (R3). Circulation 1998;98:I-17. Le Breton H, Plow EF, Topol EJ. Role of platelets in restenosis after percutaneous coronary revascularization. J Am Coll Cardiol 1996;28:1643-51. Simon DI, Xu H, Ortlepp S, Rogers C, Rao NK. 7E3 monoclonal antibody directed against the platelet glycoprotein IIb/IIIa crossreacts with the leukocyte integrin Mac-1 and blocks adhesion to fibrinogen and ICAM-1. Arterioscler Thromb Vasc Biol 1997; 17:528-35. Gould RJ, Polokoff MA, Friedman PA, Huang TF, Holt JC, Cook JJ, et al. Disintegrins: a family of integrin inhibitory proteins from viper venoms. Proc Soc Exp Biol Med 1990;195:168-71. Niewiarowski S, McLane MA, Kloczewiak M, Stewart GL. Disintegrins and other naturally occurring antagonists of platelet fibrinogen receptors. Sem Hematol 1994:31:289-300. Scarborough RM, Rose JW, Naughton MA, Phillips DR, Nannizzi L, Arfsten A, et al. Characterization of the integrin specificities of disintegrins isolated from American pit viper venoms. J Biol Chem 1993;268:1058-65. Scarborough RM, Rose JW, Hsu MA, Phillips DR, Fried VA, Campbell AM, et al. Barbourin: a GPIIb/IIIa specific integrin antagonist from the venom of Sistrurus barbouri. J Biol Chem 1991;266:9359-62. Scarborough RM, Naughton MA, Teng W, Rose JW, Phillips DR, Nannizzi L, et al. Design of potent and specific integrin antagonists: peptide antagonists with high specificity for glycoprotein IIb/IIIa. J Biol Chem 1993;268:1066-73.

American Heart Journal December 1999

27. Phillips DR, Scarborough RM. Clinical pharmacology of eptifibatide. Am J Cardiol 1997;80(suppl 4A):11B-20B. 28. Scarborough RM. Eptifibatide. Drugs Future 1998;23:1-6. 29. Phillips DR, Teng W, Arfsten A, Nannizzi-Alaimo L, White MM, Longhurst C, et al. Effect of Ca2+ on GP IIb/IIIa interactions with integrilin: enhanced GP IIb/IIIa binding and inhibition of platelet aggregation by reductions in the concentration of ionized calcium in plasma anticoagulated with citrate. Circulation 1997;96:1488-94. 30. Harrington RA, Kleiman NS, Kottke-Marchant K, Lincoff AM, Tcheng JE, Sigmon KN, et al. Immediate and reversible platelet inhibition after intravenous administration of a peptide glycoprotein IIb/IIIa inhibitor during percutaneous coronary intervention. Am J Cardiol 1995;76:1222-7. 31. Uthoff K, Zehr KJ, Geerling R, Herskowitz A, Cameron DE, Reitz BA. Inhibition of platelet adhesion during cardiopulmonary bypass reduces postoperative bleeding. Circulation 1994;90[part 2]:II269-II-74. 32. Charo IF, Scarborough RM, du Mee CP, Wolf D, Phillips DR, Swift RL. Pharmacodynamics of the GP IIb/IIIa antagonist Integrelin: phase I clinical studies in normal healthy volunteers [abstract]. Circulation 1992;86(suppl I):I-260. 33. Ohman EM, Harrington RA, Lincoff AM, Kitt MM, Kleiman NS, Tcheng JE. Early clinical experience with integrelin, an inhibitor of the platelet glycoprotein IIb/IIIa integrin receptor. Eur Heart J 1995;16(suppl L):50-5. 34. Schulman SP, Goldschmidt-Clermont PJ, Topol EJ, Califf RM, Navetta FI, Willerson JT, et al. Effects of Integrelin, a platelet glycoprotein IIb/IIIa receptor antagonist, in unstable angina: a randomized multicenter trial. Circulation 1996;94:2083-9. 35. Ohman EM, Kleiman NS, Gacioch G, Worley SJ, Navetta FI, Talley JD, et al. Combined accelerated tissue-plasminogen activator and platelet glycoprotein IIb/IIIa integrin receptor blockade with Integrilin in acute myocardial infarction: results of a randomized, placebo-controlled, dose-ranging trial. Circulation 1997;95:846-54. 36. Tcheng JE, Harrington RA, Kottke-Marchant K, Kleiman NS, Ellis SG, Kereiakes DJ, et al. Multicenter, randomized, double-blind, placebo-controlled trial of the platelet integrin glycoprotein IIb/IIIa blocker Integrelin in elective coronary intervention. Circulation 1995;91:2151-7. 37. The IMPACT-II Investigators. Randomised placebo-controlled trial of effect of eptifibatide on complications of percutaneous coronary intervention: IMPACT-II. Lancet 1997;349:1422-8. 38. Tcheng JE. Impact of eptifibatide on early ischemic events in acute ischemic coronary syndromes: a review of the IMPACT II trial. Am J Cardiol 1997;80(suppl 4A):21B-8B. 39. Harrington RA. Design and methodology of the PURSUIT trial: evaluating eptifibatide for acute ischemic coronary syndromes. Am J Cardiol 1997;80(suppl 4A):34B-8B. 40. Harrington RA, on behalf of the PURSUIT Trial Investigators. Inhibition of platelet glycoprotein IIb/IIIa with eptifibatide in patients with acute coronary syndromes. N Engl J Med 1998;339:436-43. 41. Talley JD. News and views: progress in interventional cardiology. J Interv Cardiol 1997;10:387-90.